On the convergence of rank-one multi-target linear regression

This paper presents a new method for solving rank-one multivariate regression problems, providing solution that maximizes the sum of squared correlations one-dimensional fitted pattern with target variates. The suitability and consistency estimator are formally proved experimentally tested. In particular, it is shown estimate converges not only as function number items, but also An equivalent conventional reduced-rank case identified, inherits convergence properties approach. Application programs in Matlab/Octave code provided numerical examples using artificial data well real presented.